Across the Chemical Engineering & Materials Science department, computational modeling and artificial intelligence are used as tools to accelerate materials and molecule discovery, replace or complement expensive experimental characterization, and predict properties across multiple length and time scales. Common themes include machine-learned interatomic potentials that bridge quantum mechanical accuracy with the speed needed to model reactions and materials at scale; multiphysics and microstructure-level simulations that link atomic-scale structure to device-level performance in batteries, fuel cells, and structural metals; and the integration of machine learning with quantum chemistry, molecular dynamics, and spectroscopy to build predictive thermodynamic and transport models. A parallel thread applies computational and AI methods — including graph neural networks and protein docking simulations — to biomolecular and protein engineering problems, while high-throughput and automated screening approaches are increasingly used to search large design spaces for optimal materials formulations.
